✨ About The Role
- The role involves researching and developing new methods for training models to tackle difficult reasoning problems that require extensive thought processes.
- Interns will explore scalable oversight protocols to enhance human capabilities in producing and quality controlling reasoning chains.
- The position includes investigating strategies to refine data pipelines for model improvement.
- Interns will study the generalization boundaries of reasoning models to inform advancements driven by data.
- The development of new benchmarks or evaluation methods for assessing advanced reasoning capabilities is a key responsibility.
âš¡ Requirements
- Candidates should be currently enrolled in a BS/MS/PhD program with a focus on Machine Learning, Deep Learning, Natural Language Processing, or Computer Vision.
- A strong background in developing models for complex reasoning tasks, such as expert-level math or software engineering, is highly encouraged.
- Prior work experience or research publications in related fields, particularly on large language models (LLMs) and reasoning, is essential.
- Proficiency in training neural networks using Python is required for this role.
- Fluency in English, both spoken and written, is necessary for effective communication within the team.